Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
amsha
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
15 ms
·
1.
▲
by
amsha
6mo ago
Yeah, the underlying sandbox technology between Ash and CC is fundamentally different. Ash is built on the Endpoint Security and Network Extension APIs. Together, they cover the full gamut of potential sandbox escapes, and it's a simpl
2.
▲
by
amsha
6mo ago
Good idea about the GitHub stub repo, I've added it here: https://github.com/Ash-Sandbox/bugs . Your feedback about permissions is something I'm working on fixing now. Apple requires multiple permissions for E
3.
▲
by
amsha
6mo ago
Yeah I think that's a good point. I've added process-specific network access to the roadmap: https://github.com/Ash-Sandbox/bugs/issues/1
4.
▲
by
amsha
6mo ago
I'd be interested in seeing a breakdown in which ones use: - VMs - Containers - sandbox-exec (macOS builtin tool) - Endpoint Security + Network Extension (AFAIK this is just Ash but it would be good to see company here)
5.
▲
by
amsha
6mo ago
Yeah I like the idea of allowing the sandbox to restrict network connections to specific processes. I'll put it on the roadmap: https://github.com/Ash-Sandbox/bugs/issues/1
6.
▲
Show HN: Ash, an Agent Sandbox for Mac
(ashell.dev)
16 points
by
amsha
6mo ago
|
15 comments
7.
▲
by
amsha
1y ago
There is a very to-the-point diagram called the Hydrogen Ladder [0] that classifies the usefulness of hydrogen by domain. The author makes a good case that, despite its versatility, hydrogen is almost always worse than some other clean tech
8.
▲
by
amsha
5y ago
Email federation is more than just the addressing system. Even with a consistent address, your host still needs to be able to send and receive SMTP messages to participate in the ecosystem. Address lock-in is real but doesn't fully neg
9.
▲
by
amsha
5y ago
One unmentioned feature of federation is that it lowers the cost of switching. If your provider begins to misbehave, you can jump to another one. This is especially true in email, where you can move from one host to another by updating your
10.
▲
by
amsha
6y ago
This looks pretty interesting. Besides telephony and other message-passing applications, what are the hero use cases for an actor system on an embedded device?
11.
▲
by
amsha
7y ago
The structure of your website's data is less important than its moderation process. High quality online communities tend to have strong, well-defined moderation policies – Wikipedia and Stack Overflow are good examples, as are HN and c
12.
▲
by
amsha
7y ago
> I think that in studying economic growth, we (and especially we in the Silicon Valley) focus way too much on gadgets, and too little on the simple fact of human knowledge of how to do things. This is a major component of software'
13.
▲
by
amsha
8y ago
It’s dynamically-typed and it doesn’t have real arrays. It can sometimes be frustratingly slow to do things that imperative languages do easily, but that’s the price of entry for functional languages. Luckily you can drop down into Rust or
14.
▲
by
amsha
8y ago
That's true. My performance claims are biased toward offline-capable editors, where the last consistent state between sites may be thousands or millions of edits ago. For online-only editors the set of edits between now and the most re
15.
▲
by
amsha
8y ago
My understanding of OT is that each change requires traveling backward though the edit history to find the first state that is consistent between sites, then traveling forward to transform the new change against past edits. In the worst cas
16.
▲
by
amsha
8y ago
Here's the key sentence: "concrete implementations of CRDT in co-editors revealed key missing steps in CRDT literature." This paper may be correct for academic CRDTs but it is very wrong when looking at industry implementatio
17.
▲
by
amsha
8y ago
I would go with the simplest solution that lets you ship. For basic web apps, I find k8s is overkill. However, once you need service discovery, secret management, auto scaling, etc k8s is the least bad solution I’ve found. I ended up using
18.
▲
by
amsha
9y ago
Are there any specific actions Seattle has taken to make the driving experience worse?
19.
▲
by
amsha
10y ago
That's what I'm currently using! The downside with configuring via CLI is that anyone who needs to update the cluster has to download the tool, make sure they have the right AWS keypair installed, make sure their config is correct
20.
▲
by
amsha
10y ago
Unfortunately ECS doesn't provide Kubernetes clusters, it's a proprietary orchestration platform with its own feature set.
21.
▲
Kubernetes cluster management for AWS?
2 points
by
amsha
10y ago
|
4 comments
22.
▲
by
amsha
10y ago
I'd mostly agree, and the packages you've written (ExRM and Distillery) make releases super easy to build, start, and stop. The main benefit of Go, I think, is that it can cross-compile builds. Erlang and Elixir don't have
23.
▲
by
amsha
10y ago
Yeah, disabling the discrete card made a huge difference to my laptop's stability (also 2010 15" MBP with NVIDIA GT 330M). There's an open-source app called gfxCardStatus [0] that lets you switch between integrated and discre
24.
▲
by
amsha
10y ago
Does anyone here have insight into how widely Pronovost's ideas have been implemented since the article was published?
25.
▲
by
amsha
10y ago
Yes, email is federated.
26.
▲
by
amsha
10y ago
Any programming language can do IPC, but Erlang/Elixir is the only language I've seen that makes it really easy. Erlang IPC is transparent whether you're sending messages between to local processes or remote processes. Most
27.
▲
by
amsha
10y ago
In my experience, writing (or any other art) is maybe the most downwardly-mobile profession there is. The supply of artists far outstrips the demand for art, and getting your first job often depends on proximity to industry gatekeepers. I
28.
▲
by
amsha
11y ago
More concretely, Phoenix sets itself apart through its own features, Elixir's features, and an (as far as I've seen) unparalleled commitment to the community: * PubSub, so if you're building a communication-based app your job
29.
▲
by
amsha
11y ago
Security and fault-tolerance are cost centers, so unless they are explicit features of a deliverable they will be ignored. Planes, cars, and banks all have anomaly detection and auditing because people demand it. They understand the risks o